Health surveys repeatedly show that females have higher rates of illness, disability days and health service utilization than males. Numerous hypotheses have been proposed to explain the sex differences. They invoke health attitudes and beliefs, social roles, social stress, interview behavior, life style behaviors, and genetic factors. No study has ever simultaneously tested these hypotheses. The main purpose of this research is to determine how social and social psychological factors affect reports of symptoms, disability, and health actions, and how these factors specify (account for) sex differences. A second purpose is to document how health symptoms are channelled into health services, self-treatment, or no health action at all. The research design and analysis strategy are: A sample of white adults in the Detroit metropolitan area will keep health diaries for four weeks. Each day, respondents record information about all symptoms experienced, their perceived severity, activity restriction and mobility limitation due to symptoms, health actions (self-treatment, medical/dental attention, preventive health care not associated with symptoms), communications with peers and kin about symptoms (lay referral), and events of the day which were especially stressful or exciting. Two interviews will occur: one before the diary period to measure the main predictors, and one at the end of the diary period principally to test the diary's reactivity (effect on respondent's health awareness and health behavior). The main dependent variables generated by the diary involve counts of symptoms, disability days, and health actions during the month. Using multiple regression, predictors will be entered sequentially, and their ability to specify the "raw" sex effect will be examined. Some important innovations of the research are: the use of a health diary among males and persons of low socioeconomic status; reports of all symptoms, not simply those which restrict activity or receive medical attention; self-response (no proxies allowed); extensive data on the activity and effectiveness of the "lay referral system"; and data on health actions other than medical/dental attention.